On Wed, Nov 10, 2021 at 10:56:42AM +0200, Andy Shevchenko wrote:
> On Wed, Nov 10, 2021 at 10:21:36AM +0200, Sakari Ailus wrote:
> > On Wed, Nov 10, 2021 at 10:18:20AM +0200, Andy Shevchenko wrote:
> > > On Wed, Nov 10, 2021 at 10:09:04AM +0200, Sakari Ailus wrote:
> > > > On Tue, Nov 09, 2021 at 02:19:13PM +0200, Andy Shevchenko wrote:
> > > > > On Tue, Nov 09, 2021 at 01:19:34PM +0200, Sakari Ailus wrote:
>
> ...
>
> > > > > > - } else if (is_acpi_device_node(fwnode)) {
> > > > > > + }
> > > > >
> > > > > > + if (is_acpi_device_node(fwnode)) {
> > > > >
> > > > > Unneeded change. Yes I know that 'else' here can be skipped. But in such cases
> > > > > it's a trade-off between changes, code readability and maintenance. Since here
> > > > > it's a fix, backporting concerns are also play role.
> > > >
> > > > The patch applies cleanly to 5.5, the oldest kernel where it's needed.
> > >
> > > Why? I don't see how this affects the workflow.
> > >
> > > > Do you prefer another patch to remove the else clause?
> > >
> > > Nope.
> > >
> > > > I think it's a bit overkill...
> > >
> > > Exactly, that's why the question is why have you split the if-else-if to
> > > two if:s?
> >
> > The else clause is useless, I think the code simply looks better without
> > it.
>
> I see a contradiction here:
>
> Statement 1: 'else' is useless.
> Statement 2: patch to remove it is overkill.
There's no contradiction.
I argue doing that in a separate patch is waste of everyone's time. As
simple as that.
Sure, it could be done, but usually ends up being left as-is.
--
Sakari Ailus
